The Psychology of Learning, 2
Psychologists discover how to reliably
elicit changes in behavior through
experience
• Classical conditioning
– Pavlov
• Operant conditioning
– Law of Effect: Thorndike
• Power of a stimulus to evoke a response is
strengthened when the response is followed by a
reward and weakened when it is not followed by a re
Psychologists discover how to reliably
elicit changes in behavior through
experience
• Classical conditioning
– Pavlov
• Operant conditioning
– Law of Effect: Thorndike
– Skinner: Methods of operant conditioning
Operant conditioning
Contiguity and frequency in
operant conditioning
• Behavior (response) precedes reward:
Contiguity
– Dog whops cupboard door
– Gets treat
• Happens often: Frequency
• Learning occurs
– Dog ruins cupboard door with repeated
whopping
Partial and Full Reinforcement
• Law, psychology, and Wilbur the English
Bulldog
– Psychologist: Intermittent reinforcement yields
stronger learning
– Lawyer: but it’s a CONTRACT!
• 100% reinforcement is better
– Who’s right?
• Partial Reinforcement Effect
– Responses acquired under intermittent
reinforcement are more difficult to extinguish
than those acquired with continuous
reinforcement
Are contiguity and frequency all
that matter?
•Contiguity and frequency: The original laws of
association
–A+B yields association
–A+B, A+B, A+B, A+B ..... strengthen association
•Contiguity and frequency instantiated in procedures
of operant and classical conditioning
–CS + US
• CS + US, CS + US, CS + US, CS + US.....
–Response + Reinforcement
• Rsp + Rf, Rsp + Rf, Rsp + Rf, Rsp + Rf, Rsp + Rf.....
Are contiguity and frequency all
that matter?
• No
• The role of contingency in learning
– One event predicts another with greater than
chance probability
Contingency in classical
conditioning
• Training: tones, then shocks
– Group 1: Random pairing
U
C C
S S
U
C C
S S
U
C C C
S S S
U
U
U
U
U
U
C
C
C C
C CC CC CC
S
S
S
S
S
S
S S
S S S
U
C C
S S
U
C C
S S
U
C C
S S
U
C C
S S
Contingency in classical
conditioning
• Training: tones, then shocks
– Group 1: Random pairing
Time----------------->
U
U
U
U
U
U
U
C
C C U
C
C C
C CC CC CC
C
C
C
S
S S
S
S
S
S
S
S S
S S S
S S S
U
C C
S S
U
C C
S S
U
C C
S S
U
C C
S S
U
C C
S S
Contingency in classical
conditioning
• Training: tones, then shocks
– Group 1: Random pairing
– Group 2: Contingent pairing
U
C C
S S
U
C C
S S
U
C C C
S S S
U
U
U
U
U
U
C
C
C C
C CC CC CC
S
S
S
S
S
S
S S
S S S
U
C C
S S
U
C C
S S
U
C C
S S
U
C C
S S
Contingency in classical
conditioning
• Training: tones, then shocks
– Group 1: Random pairing
– Group 2: Contingent pairing
U
C C
S S
U
C C
S S
U
C C C
S S S
U
U
U
U
U
U
C
C
C C
C CC CC CC
S
S
S
S
S
S
S S
S S S
U
C C
S S
U
C C
S S
U
C C
S S
U
C C
S S
Contingency in classical
conditioning
• Training: tones, then shocks
– Group 1: Random pairing
– Group 2: Contingent pairing
• Testing: tones then shocks in shuttle box
Results
Contiguity and frequency alone
are not enough
• For classical conditioning, conditioned
stimulus must reliably predict the
unconditioned stimulus
• For operant conditioning, behavior must
reliably predict the reward (the
reinforcement)
When contingency is present
• When I do this, that
happens…
• When I don’t do
this, nothing
happens
• Happy babies in
control!
When behavior and consequences
are not contingent
• Whatever I do has
nothing to do with
that...
• Unhappy babies
without control
• There is joy in mastery • Without mastery, there
is despair